Arash Sahebolamri

Results 35 comments of Arash Sahebolamri

Ah, you are absolutely right. In fact, that is exactly what we did in the [BYODS paper](https://dl.acm.org/doi/pdf/10.1145/3622840) (see section 5.6). There, we defined a `MapLattice` data type to use for...

FWIW, I've opened a similar issue on `allocator-api2`: https://github.com/zakarumych/allocator-api2/issues/20

You are right, it should be easy to get a `Lattice` impl for `Ord` types. Ideally, we should be able to have `impl Lattice for T`, but I think that...

I just checked. There is already a wrapper type `ascent::lattice::ord_lattice::OrdLattice` for this purpose.

Thanks for the PR @StarGazerM! You could argue that it would be useful to try and make `ascent_source` parameterized. Is this what this PR is doing? Otherwise I don't quite...